Freelance · R&D Data & AI Specialist

Reliable, compliant,
reproducible R&D processes

Des process R&D fiables,
conformes et reproductibles

I bridge science, technology, and regulatory requirements for R&D teams. Many teams live with processes they consider normal that could be far more reliable. I help them see the gap, then build integrated analytical environments that close it. Speed comes as a result, not at the expense of rigor.

Schedule a 30-min call → Download CV
4
Global R&D groups
L'Oréal · LVMH · Pfizer · BI
11+
Projects delivered
5+
Life science sectors
Cosmetics · Pharma · Biotech · Medtech · Agro-food
6+
Certifications
Aslane Mortreau portrait
They trusted me
L'Oréal R&I
LVMH Recherche
Boehringer Ingelheim
PFIZER
About

The bridge between science,
technology, and regulatory.

Le pont entre la science,
la technologie et le réglementaire.

I'm Aslane Mortreau, an independent specialist for R&D teams in cosmetics, pharma, biotech, medtech, and agro-food. My differentiator is triple fluency: I understand the science, the technology, and the regulatory constraints. That is what lets me design solutions that are scientifically sound, technically robust, and audit-ready.

Many teams live with analytical processes they consider normal: spreadsheets, ad hoc scripts, manual reporting loops. I often start by revealing what could be more reliable, then scope a path to fix it.

Beyond classical statistical modeling, I design and deploy machine learning solutions tailored to R&D contexts: predictive models on experimental data, AI-driven formulation tools, and intelligent systems that integrate directly into existing research workflows.

I also work at the process level: auditing existing analytical workflows, identifying bottlenecks and reproducibility gaps, and building data roadmaps that align R&D operations with long-term scientific and regulatory objectives.

Rather than delivering one-off analyses, I build integrated analytical environments: reproducible pipelines, internal tools usable by non-statisticians, and automated reporting systems that standardize methodologies and scale across projects.

View full curriculum vitae →

Je suis Aslane Mortreau, spécialiste indépendant pour les équipes R&D : cosmétique, pharma, biotech, medtech et agroalimentaire. Mon différenciateur, c'est une triple compréhension : la science, la technologie et le réglementaire. C'est ce qui me permet de concevoir des solutions scientifiquement solides, techniquement robustes et prêtes pour l'audit.

Beaucoup d'équipes vivent avec des process analytiques qu'elles croient normaux : tableurs, scripts ad hoc, boucles de reporting manuel. Mon point d'entrée, souvent, c'est de révéler ce qui pourrait être bien plus fiable, puis de cadrer un plan pour y remédier.

Au-delà de la modélisation statistique classique, je conçois et déploie des solutions de machine learning adaptées aux contextes R&D : modèles prédictifs sur données expérimentales, outils d'IA pour la formulation, et systèmes intelligents s'intégrant directement aux workflows de recherche existants.

J'interviens également au niveau des processus : audit des workflows analytiques existants, identification des goulots d'étranglement et des lacunes en reproductibilité, et construction de roadmaps data alignant les opérations R&D avec les objectifs scientifiques et réglementaires à long terme.

Plutôt que de livrer des analyses ponctuelles, je construis des environnements analytiques intégrés : pipelines reproductibles, outils internes utilisables par des non-statisticiens, et systèmes de reporting automatisés qui standardisent les méthodologies et s'adaptent à l'échelle des projets.

Voir le curriculum vitae complet →
Skills

Biostatistics & Data Analysis

Longitudinal data analysis Survival & time-to-event analysis Pharmacokinetic modeling Efficacy trial analysis Regulatory-grade statistical reporting
RSASPython

Data Engineering & Pipelines

Reproducible data pipelines Automated data quality monitoring Cloud data infrastructure Clinical data standardization (CDISC)
DagsterdbtDockerBigQuery · GCPAirflow

Analytical App Development

Interactive analysis applications Automated statistical reporting Decision-support tools for non-statisticians Scientific dashboards
R ShinyPython (Streamlit)

AI & Machine Learning

Supervised learning Graph machine learning NLP / text mining Computer vision Unsupervised learning
PyTorchscikit-learnHugging FaceRDKitLLMMLflow

R&D Process & Strategy

Analytical workflow audit Data roadmap design Reproducibility frameworks Cross-functional scientific collaboration
Services

What you can hire me for

Structured engagements, from a two-week audit to a full analytical platform build. Each offer has a clear scope, deliverable, and timeline.

01

Workflow audit & scoping

Map your current analytical processes, identify blind spots, reproducibility gaps, and regulatory risks. Leave with a prioritized roadmap and a clear go/no-go on next steps.

Deliverable: audit report + roadmap · 1–3 weeks

02

Proof of concept

Validate a hypothesis on your data before committing to a full build: statistical approach, ML model, pipeline prototype, or internal tool mock-up with real inputs.

Deliverable: working prototype + feasibility report · 2–6 weeks

03

Application or pipeline build

Design and deliver a production-ready analytical application (Shiny, Dash, Streamlit) or a reproducible data pipeline (CDISC, PK/NCA, quality monitoring) integrated into your environment.

Deliverable: deployed tool or pipeline + documentation · 1–4 months

04

Ongoing support

Embedded support for R&D or data teams: maintain analytical tools, extend pipelines, handle regulatory updates, and act as the bridge between scientists, IT, and quality.

Deliverable: retainer · monthly

05

Training & knowledge transfer

Upskill your team on the tools and methods you now rely on: reproducible pipelines, statistical workflows, Shiny apps, or GxP-aware data science practices.

Deliverable: workshop or hands-on sessions · 1–5 days

Experience

Professional background

May 2026 – Present

L'Oréal R&I

AI Engineer

Regulatory intelligence, unstructured data & AI validation

  • Acquire and structure unstructured data from regulatory sources, scientific literature, and related documentation
  • Build and improve LLM-powered agentic systems for regulatory intelligence, claims discovery, and scientific reasoning
  • Develop knowledge graphs for structured querying and evidence linking; contribute to evaluation frameworks and Python packages on GCP/Azure
PythonGCPBigQueryAzureLLMLangGraphNLPKnowledge GraphsDockerGitHub Actions

Mar 2026 – Sep 2026

Pfizer

Freelance Data & AI Specialist

Medical Affairs

  • Designed a standardized framework to measure and compare medical impact across multiple markets
  • Enabled transition from fragmented manual reporting to a structured and scalable analytics approach
  • Provided strategic foundations for more data-driven and predictive decision-making in Medical Affairs
Medical AffairsData StrategyKPI FrameworkAnalyticsDecision Support

Nov 2025 – Present

Al-Gebrax

Freelance R Shiny Developer

Pharmaceutical & CMC Analytics

  • Statistical analyses for pharma and CMC studies: stability, assay performance, bioanalytical workflows
  • R Shiny applications automating standardised analyses with PDF/Word report generation
  • Reproducible pipelines and data quality monitoring for regulatory-driven analytics
R · ShinyDockerCMC

Nov 2025 – Present

Gencovery

Data Science & Bioinformatics Consultant

Life Science Workflows

  • Develop Reflex bricks for life science workflows: PK-NCA, CDISC validation, molecular embeddings
  • Generic pipelines for omics data exploration, clustering, and differential expression
PythonReflexStreamlitRDKitDocker

Nov 2025 – Jan 2026

L'Oréal R&I

Freelance Data Scientist

Research & Innovation

  • Full scientific and technical feasibility study for a strategic R&I initiative
  • Delivered structured recommendations to support go/no-go decision-making at innovation leadership level
PythonChemoinformaticsData EngineeringMachine LearningR&D workflowsRegulatory

Aug 2023 – Sep 2025

LVMH Recherche

Data Research Engineer

  • Led statistical analyses of in vivo cosmetic efficacy studies: linear mixed-effects models, Kaplan-Meier, Cox regression
  • Wrote and validated Statistical Analysis Plans with clinical and regulatory teams
  • End-to-end automated statistical workflows reducing analysis turnaround time by over 50%
  • AI-driven molecular substitution engine using graph embeddings (metapath2vec)
RBiostatisticsPythonGCPData EngineeringChemoinformaticsProject Management

Oct 2024 – May 2025

Oltega

Freelance Automation Specialist

  • Automation solutions across CRM management and business workflows
  • Integrations with Make, Zapier, Monday.com, and HubSpot
PythonMakeZapierAutomationCRM
Portfolio

Selected projects

Projets sélectionnés

Biotechnology · AI / ML

PSAP

Reproducible pipeline to evaluate peptide candidates from literature through structure prediction, molecular dynamics, and AI-assisted scientific analysis.

ESMFoldGROMACSPython
View →

Biotechnology

DNA-Based Data Storage

R&D pipeline to encode digital data into DNA sequences with biochemical constraints and error correction.

PythonBioinformatics
View →

Statistics

TrialLytics

Clinical statistics automation platform with ANOVA, mixed models, survival analysis, diagnostics and automated reporting.

R · Shinylme4
View →

Statistics

PKnalytics

PK/NCA and bioequivalence platform with a full pipeline: ingestion, estimation, diagnostics, and reporting.

Python · Dash
View →

AI / ML

Raw Material Substitution

Graph model (metapath2vec) to suggest compatible substitutes for cosmetic formulations.

Graph MLPython
View →

Statistics

Automated SAS Code Generation

Automatic SAS code generation for statistical analysis of questionnaires and cosmetic claim substantiation.

SAS
View →

Data Engineering

CDISC Clinical Data Pipeline

Automated CDISC SDTM/ADaM pipeline orchestrated with Dagster for reproducible clinical data processing.

DagsterPython
View →

Data Engineering

Constellab · Great Expectations

GWS brick enabling Great Expectations validation workflows and automatic publication of Data Docs inside the platform.

PythonGWSGreat Expectations
View →

Data Engineering

CDISC Validator

SDTM/ADaM validation engine with structural, relational checks and detailed anomaly reports.

PythonCDISC
View →

Data Engineering

Random Walk Pipeline

Dockerized streaming and visualization pipeline to simulate, process, and analyze random walk data in real-time.

DockerStreaming
View →

AI / ML

AVM Detection on MRI

CNN-based detection of arteriovenous malformations in brain MRI scans.

PyTorchCNN
View →

AI / ML

Evaluo

AI/NLP application that extracts skills from CVs and generates structured competency dossiers automatically.

NLPPython
View →

AI / ML

Epidemiological Simulator

Interactive SIR simulator to evaluate testing and vaccination strategy impacts on epidemic trajectories.

PythonSimulation
View →
Testimonials

What clients say

I have worked with Aslane on several missions: impeccable rigor, professionalism, and outstanding support and advice. Highly professional from start to finish. I strongly recommend him.

Enéa Audouin

Co-founder · Oltega

Aslane quickly built strong expertise on specific and complex business topics, grasped the business stakes, and translated them into relevant, structured, and actionable data analyses. Beyond his solid technical skills, I particularly valued his autonomy, his ability to propose solutions, his analytical rigor, and his capacity to communicate effectively with senior stakeholders. Aslane was a real asset in securing the project's feasibility under tight deadlines.

Océane Doublet

Data and AI Department Head · Enovalife

I recommend Aslane for data engineering and R-based data analysis missions. He masters data preparation, transformation, and structuring. Thanks to him, we developed R Shiny analyses and applications integrated into cloud environments (AWS/GCP), connected to data pipelines and storage. His deliverables are robust, clear, and directly usable by business teams.

Sofiane Djerbi

Senior DevOps Engineer

He is a colleague who is at once very competent, intelligent, and pleasant: collaboration with him is smooth and effective. He builds expertise quickly in his areas, works well in a team, and his contributions are always relevant and well thought out. A real added value in his field.

Laurie Montjoly

Toxicology & Ecotoxicology Scientist · Data and AI

Aslane proved serious, committed, and reliable throughout the collaboration. He adapted to project constraints, worked in a structured way, and maintained clear, professional communication. The collaboration went very well and the work delivered was of high quality.

Mohamed Lakhdar

Founder & Director · Al-GebraX

Blog

Insights on R&D data

Statistics 2025

Why mixed-effects models matter for your cosmetic efficacy trials

Most R&D teams still use basic ANOVA on repeated-measures data. Here's what they're missing.

Read more →
Data Engineering 2025

How to build a reproducible R&D data pipeline from scratch

Excel and ad-hoc scripts are not a pipeline. A practical guide to building something that scales.

Read more →
Regulatory 2025

The GxP data science stack for a 5-person biotech team

S3, dbt, renv, signed Git tags, Quarto. A lean, audit-ready architecture for small teams who cannot afford to get compliance wrong.

Read more →
View all articles →
Contact

Let's work together

Looking for a specialist who speaks science, tech, and regulatory? Whether it's a workflow audit, a scoped POC, or a full analytical platform: let's talk.

Phone

+33 6 27 66 05 07

Ready to make your R&D processes more reliable?

From workflow audits to reproducible pipelines and regulatory-grade reporting: I help life science teams build analytical environments they can trust, and scale with confidence.

Book a free 30-min call → Send an email → Envoyer un email →